All ON pathways are not alike

ON-type bipolar cells (ON-BPCs) in the retina relay the detection of increases in light intensity from rod and cone photoreceptors, the primary sensory neurons of the visual system, to the ganglion cells, the output neurons of the retina. The light response of ON-BPCs is mediated by metabotropic glutamate receptor 6 (mGluR6), a G protein-coupled receptor (GPCR) on the ON-BPC dendrites that is activated by glutamate, the neurotransmitter released by photoreceptors. Glutamate release is high in darkness and drops in response to light-induced hyperpolarization of the photoreceptors. In the dark, tonic activation of the mGluR6 pathway maintains the transient receptor potential cation channel subfamily M member 1 (TRPM1) cation channel in a closed state. The light response of ON-BPCs is initiated by the inactivation of mGluR6, leading to opening of the TRPM1 channels and depolarization of the cell (Fig.1).

Figure 1.

How light causes depolarization of ON-BPCs

In the dark, mGluR6 binds to glutamate and activates Go. The alpha subunit of Go exchanges GDP for GTP and dissociates from the beta and gamma subunits. The βγ heterodimer maintains the TRPM1 channel in a closed state. In the light, the glutamate concentration in the synaptic cleft falls, inactivating mGluR6 and Go. Goα hydrolyses GTP to GDP, a process that is accelerated through interaction of Goα with RGS7 and RGS11. GDP-bound Goα recombines with the βγ subunits. TRPM1 opens and the inward current depolarizes the cell.

The heterotrimeric G protein linking mGluR6 activity to the gating of TRPM1 is Go. As in all GPCR pathways, activation of the receptor triggers the exchange of GDP for GTP by the G protein alpha subunit and the dissociation of Goα-GTP from Gβγ. Results from Shen et al. (2012) indicate that it is Gβγ rather than Goα that closes TRPM1, and thus the identity of the specific Gβ and Gγ subunits is of great interest. The Gβ subunit is likely to be Gβ3: immunostaining for Gβ3 reveals strong immunoreactivity in ON-BPCs, knockout of Gβ3 in mice severely attenuates the ON-BPC component of the electroretinogram, and light responses from individual ON-BPCs are greatly reduced (although not eliminated) in the Gβ3 knockout (KO) animals (Dhingra et al. 2012). Genetic deletion of Gβ3 also caused a reduction in expression of other components of the mGluR6 signal transduction pathway, including mGluR6 and TRPM1.

The identity of the gamma subunit was thought to be Gγ13 based on strong and specific immunostaining of all ON-BPCs, and the near absence of Gγ13 staining of ON-BPCs in the Gβ3 KO retina. Therefore, the findings of Ramakrishnan et al., in this issue of Journal of Physiology come as a surprise. The authors generated a Gγ13 KO mouse line, and discovered that the light responses of the rod ON-BPCs were reduced compared to wild-type (although not to the extent of the Gβ3 KO), whereas the light responses of cone ON-BPCs were barely affected. Given the conservation of other components of the mGluR6 pathway between rod and cone ON-BPCs, this dramatic difference is unanticipated, and begs the questions what is the Gγ subunit required for cone ON-BPC light responses and what is the role of Gγ13 in these cells?

It may be that in cone ON-BPCs there are two Gγ subunits and that knocking out only one is insufficient to cause a deficit. The light response of rod ON-BPCs is reduced in the Gγ13 KO, but only by ∼50%, suggesting that in these cells, too, there might be a second Gγ subunit in the pathway. An analogous situation exists with regard to the regulator of G protein signalling (RGS) proteins of ON-BPCs. The kinetics of the ON-BPC light response depend on the rate of GTP hydrolysis by Goα, a process that is greatly accelerated by the action of RGS proteins. There are two in ON-BPCs, RGS7 and RGS11, and knockout of either one or the other has little effect on the ON-BPC light response, but knockout of both eliminates the flash response of ON-BPCs (Cao et al. 2012). Ramakrishnan et al. (2014) identified several additional Gγ candidates expressed by ON-BPCs, including Gγ5, Gγ10 and Gγ11, but it remains to be determined whether any of these function in the mGluR6–Go–TRPM1 pathway and under what conditions.

The visual system operates over a vast range of luminance, transitioning between different cell types and circuits for optimal performance. The classic example is the dominance of the high sensitivity, low acuity rod-driven pathway in very dim light and low sensitivity, high acuity cone-driven vision in daylight. Recently, Szikra et al. (2014) have shown that rod to rod-BPC transmission can also influence vision in very bright light when rod outer segments are saturated, through cone-driven, feedforward inhibition from horizontal cells that are synaptically connected to both rods and cones. Under these conditions, the rod ON-BPCs hyperpolarize in response to increases in light intensity, and transmit surround inhibition to the inner retina. It is becoming increasingly apparent that ON-BPCs modulate their responses to serve different visual functions depending on luminance conditions. It is likely that seemingly redundant proteins, such as multiple RGS or Gβ and Gγ proteins within the ON-BPC, serve both overlapping and discrete signal transduction pathways that regulate these different functional modalities and the transitions between them. Physiological experiments testing ON-BPC light responses in the intact retina over a range of background light intensities will be necessary to differentiate the roles of these proteins.